Since the majority of hepatocellular carcinoma (HCC) patients suffer from a co-existing predisposing chronic liver disease (CLD), risk stratification of HCC patients is defined as the ability to predict outcomes from a given intervention (such as systemic therapy) by arranging patients according to the severity of their underlying liver disease. Child-Pugh (CP) score is the current standard tool for assessing the underlying CLD status in clinical practice and stratifying patients in HCC therapeutic trials. Moreover, it is an essential parameter in most currently used HCC staging systems. Multiple expert panels consensus statements concluded that patients with HCC must have a CP score of A to be considered for systemic therapies in practice or in clinical trials. This selection criterion facilitates assessment of the effect of treatment without the confounding issues of liver failure and death as a result of underlying CLD. CP score uses five variables: three objective laboratory-based (serum albumin, prothrombin time, bilirubin), in addition to two subjective clinical parameters (hepatic encephalopathy and ascites). The last two parameters are clinically difficult to grade and may vary in severity in response to medical therapies. Therefore, CP fails to accurately provide confident prediction of survival and treatment adverse events in HCC patients. Thus, major refinement of the CP score is critically needed. Circulating levels of insulin-like growth factor-1 (IGF-1) decrease sharply in patients with CLD including cirrhosis and HCC, because the liver is responsible for synthesis of most of the circulating IGF-1. Most recently, we published (Kaseb et al, JNCI 2014) a paper describing the development and validation of an innovative plasma IGF score by replacing clinical assessment of ascites and encephalopathy in the CP score with the objectively quantified plasma IGF-1. Notably, a significant number of HCC patients with old CP-A class were reclassified into CP-B (26% in the testing set, and 46% in the validation set). Patients categorized as old A/new B had a significantly poorer OS than did patients categorized as old A/new A in both the training and validation cohorts (P = 0.026 and <0.001, respectively). A major advantage to the use of plasma IGF-1 is being a well-characterized existing plasma assay that is ready for implementation. We hypothesize that among patients with old CP-A class HCC treated with sorafenib, those who are reclassified as new CP-B will have shorter overall survival and time-to-tumor-progression and a higher rate of adverse events than will those reclassified as new CP-A. We will test this hypothesis by conducting a prospective double- blinded biomarker study to compare the prognostic stratification ability of the conventional 'old' Child-Pugh score to the 'new' IGF score in 100 HCC patients (all Child-Pugh A) to be treated with sorafenib systemic therapy. Thus, our study is designed to reveal the potential superiority of the new score over the old CP.